The regulator hissed, a dry, metallic gasp that always feels a bit too personal when you are 8 feet under the surface of a saltwater tank. I was scrubbing a patch of stubborn brown algae off a decorative coral head, the kind of repetitive, mindless task that allows your brain to drift into the middle distance while your hands do the heavy lifting. Down here, the physics are honest. If the filter is clogged, the water gets cloudy. If you don’t scrub the glass, the tourists can’t see the sharks. There is no ‘alignment meeting’ for a clogged intake valve. You either fix it, or the ecosystem fails.

The Meeting: Optimizing the Description of Work

We sat in that room for 48 minutes. We were there to discuss a new software suite that would allow us to ‘visualize our scrub-frequency metrics.’ There were 8 of us in the room, including three people whose job titles involved the word ‘strategy’ but who I am fairly certain have never touched a wet fish in their lives. We moved tickets from one digital column to another. We assigned ‘story points’ to the act of cleaning the filtration array. It was a masterpiece of meta-work. We were optimizing the description of the work, the reporting of the work, and the feeling of the work, without actually touching the water.

The Flat-Pack Hallucination

This is the state of the modern world. We have become world-class experts at building the scaffolding while forgetting to lay a single brick. Last weekend, I tried to assemble a new sideboard for my living room. It was one of those flat-pack nightmares that promises ‘Scandi-minimalism’ but delivers ‘Instruction-induced psychosis.’ Halfway through, I realized there were 18 missing bolts. Not one, not two, but nearly twenty critical pieces of structural integrity were simply absent.

What did I do? Instead of calling the manufacturer or driving to the hardware store, I spent three hours organizing the remaining screws by length and color. I even made a little chart on a piece of cardboard. I optimized my workspace. I cleared the floor. I made the process of building the sideboard look like a surgical operation. But at the end of the day, I still didn’t have a sideboard. I had a heap of laminate and a very organized pile of useless hardware. I was avoiding the fundamental brokenness of the situation by obsessing over the parts I could control.

We do this in tech every single day. We have 15 different project management tools, 8 communication channels, and a ‘sprint cadence’ that would make an Olympic track coach weep with joy. And yet, it still takes 68 days to get a simple data extract from the legacy system. We spend two hours in a planning meeting, assigning complexity scores to a Jira ticket that says ‘Extract user data.’ We debate whether it’s a ‘3’ or an ‘8’ in terms of difficulty. We talk about the ‘deliverables’ and the ‘stakeholder expectations.’

But the actual task? The actual work of getting that data? It’s going to take one engineer 108 hours of miserable, soul-crushing labor because the three legacy databases don’t talk to each other, the documentation was written by someone who left the company in 2008, and the API is essentially a series of mirrors and smoke signals. No amount of Trello boards will change the fact that the core task is fundamentally broken. We are trying to polish a stone that is actually a lump of coal, and we’re wondering why our hands are still dirty.

The Shovel vs. The Spreadsheet

This obsession with the meta-work-the administration of the task rather than the task itself-is a classic avoidance tactic. It’s a collective hallucination that allows us to feel productive without having to face the terrifying, messy reality of technical debt. It is much easier to hold a ‘post-mortem’ about a failed data migration than it is to actually rewrite the fragile, 18-year-old script that caused the failure in the first place. One requires a spreadsheet; the other requires a shovel.

I see this in the aquarium, too. Management once spent $8,888 on a consultant to improve ‘cross-departmental communication flow.’ We had workshops. We had stickers. We had a new app where I could ‘tag’ the life-support team if a pump failed. But the pumps were still the same ones we’d been nursing along since 1998. The ‘communication flow’ was perfect, but the actual water flow was dangerously low. We were optimizing the signal, but the pipe was still full of gunk.

– Aquarium Technician

Real productivity isn’t about how many tickets you closed this week; it’s about whether the core problem actually got smaller. If you are a data scientist and you spend 88% of your time cleaning CSV files because the ingestion pipeline is a disaster, you aren’t being productive. You are just being a very expensive janitor for a system that shouldn’t be making that much mess in the first place.

This is why I find the approach of

Datamam

so refreshing in a world of fluff. They don’t give you a new dashboard to track how bad your data is; they go into the basement and fix the pipes. They understand that the ‘meta-work’ of managing bad data is a tax that companies shouldn’t have to pay. When the core task-the data extraction, the pipeline, the actual movement of information-is handled correctly, 90% of the meetings about ‘data integrity’ simply vanish. You don’t need a status update for something that just works.

Comfort Blankets and False Security

I often think about those 18 missing screws. If I had spent those three hours just finding a way to secure the boards together-maybe with some old-fashioned wood glue or some mismatched bolts from the junk drawer-I would have a place to put my books. Instead, I had a beautiful, optimized, ‘Agile’ pile of trash. We are so afraid of the ‘unclean’ work-the manual intervention, the deep-dive into the messy code, the physical reality of the problem-that we retreat into the safety of the process.

Process is a comfort blanket. It gives us a vocabulary to describe our failure as a ‘blocker’ or a ‘dependency.’ It turns a disaster into a data point. But in the aquarium, if the oxygen levels drop, the fish don’t care about your Jira velocity. They don’t care that the ticket was moved to ‘In Progress’ three days ago. They need the air.

We need to start asking ourselves: Is this meeting helping me solve the problem, or is it helping me forget that the problem is unsolvable with my current tools? Are we optimizing the work, or are we just optimizing our avoidance of the work?

The Clear Water of Direct Action

Back in the tank, the work is simple. Scrub. Check the levels. Feed. It’s hard, it’s wet, and it’s often gross. But at the end of the shift, the water is clear. You can see the 8-foot sand tiger shark gliding through the blue, and you know exactly what you did. There is no ambiguity. There is no ‘alignment.’ There is just the work, and the result.

Maybe we need more divers in the office. People who aren’t afraid to get their suits wet and ignore the ‘sprint planning’ if the filters are failing. We need to stop rewarding the people who create the most beautiful charts and start rewarding the people who actually fix the broken things at the bottom of the stack. Because eventually, no matter how many tickets you move, the water is going to get too cloudy to see through, and all your ‘status updates’ won’t mean a thing when the sharks start hitting the glass.

Order Over Outcome

I still haven’t finished that sideboard. It’s sitting in my hallway, a monument to my own desire for order over outcome. I think about it every time I walk past. It’s a reminder that the most efficient process in the world is useless if you’re missing the pieces that actually hold the thing together. We are so busy being ‘efficient’ that we’ve forgotten how to be effective.

Next time you’re in a room with 18 other people, discussing a ‘roadmap’ for a task that everyone knows is impossible because of a foundational flaw you’re all ignoring, try something radical. Stop talking about the process. Stop looking at the Jira board. Point at the broken thing-the legacy system, the manual data entry, the missing bolts-and suggest that maybe, just maybe, you should go get a shovel.

It might be messy. It might not fit into a ‘sprint.’ But at least you’ll be doing the work, rather than just talking about how you’re going to talk about it next week at 10:08 AM.